Files
ComfyUI-Lora-Manager/py/services/websocket_manager.py
T
Will Miao d572292142 feat(download): fill model metadata from the source API on download
A ModelScope or Hugging Face download landed as a bare filename, hash and
source link; the model card stayed empty until the user ran "Enrich
Metadata with AI" by hand. But everything that makes a CivitAI download
useful — the display name, the description, the tags, the trigger words,
the example images, the preview — is already published by those sites'
public APIs, so asking for it at download time is deterministic work, not
model work.

Add `py/services/model_sources/hydration.py`, called by
`_save_source_metadata()` once the sidecar exists and the file is in the
scanner cache. It fetches the model card plus the site's card extras and
hands them to the same `PostProcessor` the AI skill uses, with an empty
`llm_output`, so the two paths cannot drift apart. What lands:

* `model_name` from the site's own display name (ModelScope's `Name`), so
  the card stops showing the local filename — written only while the value
  still equals the file stem, since once a user renames a model that
  choice is theirs to keep
* `civitai.name` from the matched version's label (`showName`), which the
  card renders as the version chip
* `civitai.description` / `modelDescription` from the author summary plus
  the README as HTML
* `civitai.images` / `preview_url` from the per-file example images
* `civitai.trainedWords` from the per-file trigger words
* `base_model`, `tags` and `usage_tips` as before

Provenance stays honest: the pass records
`metadata_source = "source:<platform>"` rather than the skill's
`agent:enrich_hf_metadata`, and — because no provider ran — it no longer
stamps `llm_enriched_at`; that stamp is now conditional on the LLM
actually answering, which is what the field means. The five hand-rolled
`civitai` dict merges in the post-processor collapse into one
`_merge_civitai()` helper.

Two guards keep it safe. Only a model whose stored
`source_platform`/`source_url` match the repository being downloaded is
updated, so a local file that merely shares a name never receives another
model's card; and a file already on disk is topped up too, which
back-fills models downloaded before this existed. READMEs and detail
payloads describe the repository rather than the file, so a short-lived
process-wide `ModelSourceCache` (300 s, 32 entries) keeps a batch over one
repository to two HTTP requests. Every failure is logged and swallowed:
hydration can never fail a download.

Fix the hash policy while here. `_save_source_metadata()` went straight to
`MetadataManager.create_default_metadata()`, bypassing the per-type
factory on the owning scanner, so a checkpoint paid a full SHA256 inside
the download request — `CheckpointScanner`/`OtherScanner` deliberately
record `hash_status="pending"` with an empty `sha256` for their multi-GB
files. Metadata is now created through `scanner._create_default_metadata()`.
Hydration copes with the empty hash: `_matching_versions()` falls back to
the repository basename, which is exactly what the download just wrote.

Report both post-transfer stages, which advance no byte counter and so
read as a stall: the bar sat at 100% showing `0 B/s` for the seconds spent
hashing and fetching. `_report_phase()` broadcasts
`{"status": "metadata", "stage": "indexing" | "source", "platform": ...}`,
and `LoadingManager` names the stage in the status line (keeping the batch
position), retitles the item line, replaces the dead speed figure and runs
a sheen over the bar. `stage`/`platform` are machine-readable; the wording
is localised in the frontend.

Finally, `modelscope.ai` is its own catalogue rather than an alias of
`modelscope.cn` — `referall13/EM1` exists only on `.ai` and
`jj3550945163/Krea-2-LORA` only on `.cn` — so its URLs were rejected with
"Invalid model URL format". Register it as `ModelScopeIntlSource`
(`platform="modelscope-ai"`, `msai:` group prefix, its own default
download directory) and derive every URL either deployment builds from a
per-class `base_url`. `modelscope.com` stays an alias of `.cn`, which is
what it redirects to. The frontend source table, the link dialog hints and
the docs mirror the split.

Verified against the live APIs: both reported `.ai` repositories list
their files, read their READMEs and yield name / version / base model /
trigger words / example images. Backend 3092 passed; frontend 1259 JS +
91 Vue passed. The nine locales carry the new progress copy in the next
commit.
2026-09-17 07:47:07 +08:00

320 lines
12 KiB
Python

import logging
from aiohttp import web
from typing import Set, Dict, Optional, Any
from uuid import uuid4
import asyncio
from datetime import datetime, timedelta
logger = logging.getLogger(__name__)
class WebSocketManager:
"""Manages WebSocket connections and broadcasts"""
def __init__(self):
self._websockets: Set[web.WebSocketResponse] = set()
self._init_websockets: Set[web.WebSocketResponse] = set() # New set for initialization progress clients
self._download_websockets: Dict[str, web.WebSocketResponse] = {} # New dict for download-specific clients
# Add progress tracking dictionary
self._download_progress: Dict[str, Dict[str, Any]] = {}
# Cache last initialization progress payloads
self._last_init_progress: Dict[str, Dict[str, Any]] = {}
# Add auto-organize progress tracking
self._auto_organize_progress: Optional[Dict[str, Any]] = None
# Add recipe rematch progress tracking
self._recipe_rematch_progress: Optional[Dict[str, Any]] = None
self._auto_organize_lock = asyncio.Lock()
async def handle_connection(self, request: web.Request) -> web.WebSocketResponse:
"""Handle new WebSocket connection"""
ws = web.WebSocketResponse()
await ws.prepare(request)
self._websockets.add(ws)
try:
async for msg in ws:
if msg.type == web.WSMsgType.ERROR:
logger.error(f'WebSocket error: {ws.exception()}')
finally:
self._websockets.discard(ws)
return ws
async def handle_init_connection(self, request: web.Request) -> web.WebSocketResponse:
"""Handle new WebSocket connection for initialization progress"""
ws = web.WebSocketResponse()
await ws.prepare(request)
self._init_websockets.add(ws)
try:
await self._send_cached_init_progress(ws)
async for msg in ws:
if msg.type == web.WSMsgType.ERROR:
logger.error(f'Init WebSocket error: {ws.exception()}')
finally:
self._init_websockets.discard(ws)
return ws
async def handle_download_connection(self, request: web.Request) -> web.WebSocketResponse:
"""Handle new WebSocket connection for download progress"""
ws = web.WebSocketResponse()
await ws.prepare(request)
# Get download_id from query parameters
download_id = request.query.get('id')
if not download_id:
# Generate a new download ID if not provided
download_id = str(uuid4())
# Store the websocket with its download ID
self._download_websockets[download_id] = ws
try:
# Send the download ID back to the client
await ws.send_json({
'type': 'download_id',
'download_id': download_id
})
async for msg in ws:
if msg.type == web.WSMsgType.ERROR:
logger.error(f'Download WebSocket error: {ws.exception()}')
finally:
if download_id in self._download_websockets:
del self._download_websockets[download_id]
# Schedule cleanup of completed downloads after WebSocket disconnection
asyncio.create_task(self._delayed_cleanup(download_id))
return ws
async def _delayed_cleanup(self, download_id: str, delay_seconds: int = 300):
"""Clean up download progress after a delay (5 minutes by default)"""
await asyncio.sleep(delay_seconds)
progress_data = self._download_progress.get(download_id)
if progress_data and progress_data.get('progress', 0) >= 100:
self.cleanup_download_progress(download_id)
logger.debug(f"Delayed cleanup completed for download {download_id}")
async def broadcast(self, data: Dict[str, Any]):
"""Broadcast message to all connected clients"""
if not self._websockets:
return
for ws in self._websockets:
try:
await ws.send_json(data)
except Exception as e:
logger.error(f"Error sending progress: {e}")
async def broadcast_init_progress(self, data: Dict[str, Any]):
"""Broadcast initialization progress to connected clients"""
payload = dict(data) if data else {}
if 'stage' not in payload:
payload['stage'] = 'processing'
if 'progress' not in payload:
payload['progress'] = 0
if 'details' not in payload:
payload['details'] = 'Processing...'
key = self._get_init_progress_key(payload)
self._last_init_progress[key] = dict(payload)
if not self._init_websockets:
return
stale_clients = []
for ws in list(self._init_websockets):
try:
await ws.send_json(payload)
except Exception as e:
logger.error(f"Error sending initialization progress: {e}")
stale_clients.append(ws)
for ws in stale_clients:
self._init_websockets.discard(ws)
async def _send_cached_init_progress(self, ws: web.WebSocketResponse) -> None:
"""Send cached initialization progress payloads to a new client"""
if not self._last_init_progress:
return
for payload in list(self._last_init_progress.values()):
try:
await ws.send_json(payload)
except Exception as e:
logger.debug(f'Error sending cached initialization progress: {e}')
def _get_init_progress_key(self, data: Dict[str, Any]) -> str:
"""Return a stable key for caching initialization progress payloads"""
page_type = data.get('pageType')
if page_type:
return f'page:{page_type}'
scanner_type = data.get('scanner_type')
if scanner_type:
return f'scanner:{scanner_type}'
return 'global'
async def broadcast_download_progress(self, download_id: str, data: Dict[str, Any]):
"""Send progress update to specific download client"""
progress_entry = {
'progress': data.get('progress', 0),
'timestamp': datetime.now(),
}
for field in ('bytes_downloaded', 'total_bytes', 'bytes_per_second'):
if field in data:
progress_entry[field] = data[field]
if 'status' in data:
progress_entry['status'] = data['status']
if 'message' in data:
progress_entry['message'] = data['message']
# Post-transfer stage reporting (see `model_source_handlers._report_phase`):
# the byte counter has stopped by then, so the stage is the only thing
# that still says the download is working.
if 'stage' in data:
progress_entry['stage'] = data['stage']
if 'platform' in data:
progress_entry['platform'] = data['platform']
self._download_progress[download_id] = progress_entry
if download_id not in self._download_websockets:
logger.debug(f"No WebSocket found for download ID: {download_id}")
return
ws = self._download_websockets[download_id]
try:
await ws.send_json(data)
except Exception as e:
logger.error(f"Error sending download progress: {e}")
async def broadcast_auto_organize_progress(self, data: Dict[str, Any]):
"""Broadcast auto-organize progress to connected clients"""
# Store progress data in memory
self._auto_organize_progress = data
# Broadcast via WebSocket
await self.broadcast(data)
def get_auto_organize_progress(self) -> Optional[Dict[str, Any]]:
"""Get current auto-organize progress"""
return self._auto_organize_progress
def cleanup_auto_organize_progress(self):
"""Clear auto-organize progress data"""
self._auto_organize_progress = None
async def broadcast_recipe_rematch_progress(self, data: Dict[str, Any]):
"""Broadcast recipe rematch progress to connected clients"""
# Store progress data in memory
self._recipe_rematch_progress = data
# Broadcast via WebSocket
await self.broadcast(data)
def get_recipe_rematch_progress(self) -> Optional[Dict[str, Any]]:
"""Get current recipe rematch progress"""
return self._recipe_rematch_progress
def cleanup_recipe_rematch_progress(self):
"""Clear recipe rematch progress data if it is in a finished state"""
if self._recipe_rematch_progress and self._recipe_rematch_progress.get('status') in ['completed', 'cancelled', 'error']:
self._recipe_rematch_progress = None
def is_recipe_rematch_running(self) -> bool:
"""Check if recipe rematch is currently running"""
if not self._recipe_rematch_progress:
return False
status = self._recipe_rematch_progress.get('status')
return status in ['started', 'processing']
def is_auto_organize_running(self) -> bool:
"""Check if auto-organize is currently running"""
if not self._auto_organize_progress:
return False
status = self._auto_organize_progress.get('status')
return status in ['started', 'processing', 'cleaning']
async def get_auto_organize_lock(self):
"""Get the auto-organize lock"""
return self._auto_organize_lock
def get_download_progress(self, download_id: str) -> Optional[Dict[str, Any]]:
"""Get progress information for a specific download"""
return self._download_progress.get(download_id)
def cleanup_download_progress(self, download_id: str):
"""Remove progress info for a specific download"""
self._download_progress.pop(download_id, None)
def cleanup_old_downloads(self, max_age_hours: int = 24):
"""Clean up old download progress entries"""
cutoff_time = datetime.now() - timedelta(hours=max_age_hours)
to_remove = []
for download_id, progress_data in self._download_progress.items():
if progress_data.get('timestamp', datetime.now()) < cutoff_time:
to_remove.append(download_id)
for download_id in to_remove:
self._download_progress.pop(download_id, None)
logger.debug(f"Cleaned up old download progress for {download_id}")
async def broadcast_cache_health_warning(self, report: 'HealthReport', page_type: Optional[str] = None):
"""
Broadcast cache health warning to frontend.
Args:
report: HealthReport instance from CacheHealthMonitor
page_type: The page type (loras, checkpoints, embeddings)
"""
from .cache_health_monitor import CacheHealthStatus
# Only broadcast if there are issues
if report.status == CacheHealthStatus.HEALTHY:
return
payload = {
'type': 'cache_health_warning',
'status': report.status.value,
'message': report.message,
'pageType': page_type,
'details': {
'total': report.total_entries,
'valid': report.valid_entries,
'invalid': report.invalid_entries,
'repaired': report.repaired_entries,
'corruption_rate': f"{report.corruption_rate:.1%}",
'invalid_paths': report.invalid_paths[:5], # Limit to first 5
}
}
logger.info(
f"Broadcasting cache health warning: {report.status.value} "
f"({report.invalid_entries} invalid entries)"
)
await self.broadcast(payload)
def get_connected_clients_count(self) -> int:
"""Get number of connected clients"""
return len(self._websockets)
def get_init_clients_count(self) -> int:
"""Get number of initialization progress clients"""
return len(self._init_websockets)
def get_download_clients_count(self) -> int:
"""Get number of download progress clients"""
return len(self._download_websockets)
def generate_download_id(self) -> str:
"""Generate a unique download ID"""
return str(uuid4())
# Global instance
ws_manager = WebSocketManager()